If there’s one lesson to be learned from [Aled Cuda]’s pulsed laser driver, it’s that you can treat the current limits on electronic components as a suggestion if the current duration is measured in nanoseconds.
The components in question are a laser diode and an NPN transistor, the latter of which operates in avalanche mode to drive nanosecond-range pulses of high current through the former. A buck-boost converter brings a 12 volt power supply up to 200 volts, which then passes through a diode and into the avalanche transistor, which is triggered by an external pulse generator. On the other side of the transistor is a pulse-shaping network of resistors and capacitors, the laser diode, and a parallel array of low-value resistors, which provide a current monitor by measuring the voltage across them. There is an optoisolator to protect the pulse generator from the 200 volt lines on the circuit board, but for simplicity’s sake it was omitted from this iteration; there is some slight irony in designing your own laser driver for the sake of the budget, then controlling it with “a pulse generator we don’t mind blowing up.” We can only assume that [Aled] was confident in his work.
The video below details the assembly of the circuit board, which features some interesting details, such as the use of a transparent solder mask which makes the circuit layout clear while still helping to align components during reflow. The circuit did eventually drive the diode without destroying anything, even though the pulses were probably 30 to 40 watts. A pulse frequency of 360 hertz gave a nice visual beating effect due to small mismatches between the pulse frequency of the driver and the frame rate of the camera.
This isn’t the first laser driver to use avalanche breakdown for short, high-power pulses, but it’s always good to see new implementations. If you’re interested in further high-speed electronics, we’ve covered them in more detail before.
Ooh, an eternal pulse generator, fancy!
Is the on duration of the pulse from the “eternal pulse generator” measured in years, or much longer
“which is triggered by an eternal pulse generator”
Never seen this kind of generator in all my life.
Oops! I’ve corrected that now.
An eternal pulse is an oxymoron.
Once upon a time I was asked to develop an “LED stroboscopic inspection light” to a client.
After many trials and errors, I eventually had the idea to drive the LEDs with higher voltages, but even tho it didn’t worked in the beginning (after all, the phosphor in the white LEDs was the delaying factor), but after we started working with a mix of red, green and blue LEDs, it worked just fine, allowing an relatively strong light for a tiny bit of time, like an actual stroboscopic lamp would do.
And it worked fair longer than the original ones, too.
We pushed 85V through those LEDs for a long time haha…
The white LED phosphor doesn’t “delay”. It illuminates promptly (less than nanoseconds) when excited by blue light. Once the blue excitation stops, the phosphor has a decay time of 1-2 microseconds, with the yellow component decaying a bit faster than the red component.
The phosphor does saturate however, limiting the peak brightness you get out. Depending on the LED, the white phosphor can saturate long before the blue driving LED maxes out, making the light go blue shifted as the peak drive power goes up.
I built a LED strobe driver to drive “100 watt” white COB LEDs at up to 10 kilowatts, for durations from 100 ns to 20 us, for high-speed photography. The phosphor limits the shortest exposure to about 1 us, but what kills the project over all is the efficiency: depending on the LED (again), you get no real gain in light output past about 5-6x ‘nameplate’ current ratings, again due to the phosphor saturation. High-end LEDs (like Cree) work a bit better than the low-bidder cheapos.
As you observed, RGB arrays perform better at high power, since there’s no phosphor to saturate.
Do you have a circuit you could share?
Sorry — it’s buried in a decade old notebook. Pretty simple, no MCU: Variable HVDC source charging caps to 20-100V — the capacitor size dictates the max pulse duration. Adjustable TTL pulse generator, mosfet gate driver with lots of gate drive, and the appropriate size mosfet. Layout and conductor size suited for the risetimes and currents is needed: it doesn’t take much resistance or inductance to slow things down.
If I find the photos I’ll post to hackaday.io
Better try high pressure xenon lamp, I tried successfully soviet ISSh100 with 4500 V and 6 uF film capacitor bundle. ~2 us exposition, first tens of joules is enough for very good pictures.
Yeah, the whole project was a response to the 20 kV, 6 J lethality, deafening noise, and inefficiency of an air-gap flash. A Xenon tube is a bit of a happy medium: a bit harder to source and drive, but less lethal than air-gap and more efficient than overdriven LEDs.
But the LEDs also have a neat property of being able to be modulated in complex waveforms (Golay, Gold and similar codes) which leads to interesting applications in photoacoustics and related fields.
Schematic is here, if anyone’s interested:
https://github.com/MazinLab/BigPulse/blob/main/bigpulse/bigpulse.pdf
Many components have a “steady state” current rating and a “pulsed” current rating. As long as you stay within data sheet limits for pulse current, duration, and repetition rate, it should last a long time.
What’s interesting about this is that lasers have a very hard and definite “do not exceed” peak power: When the circulating power in the laser cavity gets high enough, its mirrors start to ablate, decreasing their reflectance, which absorbs more optical power in a runaway process. In nanoseconds it goes from “LASER” to “LED”. This will happen independent of duty cycle or average power or pulse width. The diode can be stone cold and it still will happen.
Interesting. I may be out of excuses not to try and make the visible-light blasters described on the Atomic Rockets website….
Technically an old hat, described in old and very old technical notes and publications. At Cern we used nanosecond avalanche transitor pulsers at 250V for pulsing thousands of Calorimeter channels for time calibration. Ok at that time we only had LEDs. But I guess the same works with Laser LEDs. The avalanche is like a sub-ns switch that discharges the charge Q= C*U from a capacitor onto the LED, generating instantaneous currents of tens of Amperes during typically few ns, more recently also sub-ns. In real of course it is a bit more complicated. The LED converts the larger part of the electric pulse energy into light hence not into heat which could destroy it. Surprisingly, one can see individual ns light pulses of blue LEDs in the dark by eye. This is hinting to possible applications for non-distructive optical tests at very high pulse intensities contained in single or repetitive shots.BTW, recently I built a proto of a 1ns @ 30V pulser for pulsing 250 nm UV LEDs, equivalent to photon energies of ca 5eV for detector tests. It still uses the same old principle but advanced technologies make it possible. The major issue is to get the pulse consumed by the LED and not reflected for example by its non-linear characteristics.
Ok, but for those of us who aren’t laser nerds… what does it DO?
It turns on, then off, very, very quickly. Optionally this may be repeated.
‘You can treat current limits’ should be “voltage limits”.
Nope. Though high reverse voltages and electrostatic discharge can kill a device, what kills them in normal operation is overcurrent, through heating or (in the case of lasers) too high optical output destroying the cavity mirrors. An increase in voltage drop can naturally be observed, but it’s the current that does the damage.
I looked into something like this to flash LEDs instead of Xenon flash lamps. We needed 510nm light to irradiate a Neodymium / YAG laser rod to make 1020 nm output laser light to perform oral surgery. If an LED has a max current of 50ma, but you hit it with 10A pulses for less than 1us, you can actually get away with that, and the LED continues to live. It’s not clear how many of these pulses a LED can tolerate, but they could flash many times, and we never blew one up. I was surprised. The damage to the diode is more related to the energy delivered over time. A huge amount of energy over a very short period of time does seem to stay below the damage threshold.
Something got lost in relating that story to you. Nd:YAG lasers output 1064 nm light. Often that output then gets frequency doubled to the visible 532 nm.
You can pump Nd:YAG rod with a Xenon flashlamp, but it’s pretty inefficient because Xenon has little output on the 800-900 nm range that Nd:YAG likes to absorb. Krypton flashtubes are used instead, since it has high output in that NIR range.
But you do even better by pumping with IR laser diodes around 850 nm, enabling the kHz pulse rates that you can’t do with flashlamps, and diodes are much more efficient and quieter than flashlamps. But no way would you ever pump a Nd:YAG with visible light LEDs.
We pulse blue LED’s into detectors since years without degradation. ( purpose ? we can see which channels work and align their time response at the nansosec scale). The energy released from the capacitor C at Voltage U discharged into the LED is E= 1/2 C*U**2 so for example a 22pF capacitor charged at 400V releases 1.76 microJoule ( or uWs) sounds very little but within the 5 ns pulse this corresponds to 350 Watt switched on and off. This energy does not destroy the LED because the electron-hole carriers created in the depletion layer recombine to a large percent as light (photons) and not thermally. In general one can use LEDs in a wide range of wavelenght from UV to IR but better use advanced technology like ALInGaN with very steep current increase above the forward voltage. I haven’t tested Laser LEDs so far but dont think they would get destroyed either for the same reason. Empirically it seems that organic tissues does not get damaged either, at least not by single LED pulses ( otherwise I would probably be blind). But careful with Laser LEDs , other than LEDs the Laser ones emit coherent Light which shakes up atomic bonds until they break. LED light in contrast is non coherent and averages out. But when repeating high intensity LED shots at higher rates I would also be careful. The pulse rate is limited by how fast you manage to recharge the capacitor. The power consumption is minimal, just what it needs to (re)charge C and for autonomous or insulated power an 18650 LiPo with a DCDC pump works for days.
write me when you stop the cylons with this!
Kidding aside, be interested in what kind of heat you’re introducing to the devices that are being used OFF LABEL as it were.
And how quickly said heat dissipates, in order to do it again.
Alright, next step is to rig an array of these into a beam combiner and see just how complex an engraver / cutter we can make.